Stanford University News Service



CONTACT: Stanford University News Service (650) 723-2558

Women face unique barriers to academic success in computer science

STANFORD--Eric S. Roberts had a revelation in graduate school.

The Stanford computer science professor, who is the current national president of Computer Professionals for Social Responsibility, addressed a predominantly female audience on Wednesday, Feb. 22, as part of the Jing Lyman Lecture Series on “Research on Women and Gender: 20 Years Later.”

He recalled his distress on realizing that while his male fellow students were being offered tremendous opportunities, women who were equally bright were ending up in low-level jobs in the computing industry.

Men in entry-level jobs were getting "unbelievable" salaries, Roberts said. They could control where they wanted to work, who they wanted to work for and what they wanted to do.

As women struggle through the academic pipeline from childhood to postdoctoral studies, he said, social conditioning and systematic institutional discouragement reduce their numbers in computer science.

Studies show that right after high school, almost half of students with an interest in computing are women. But women who actually receive bachelor's degrees comprise only a third of all computer scientists nationwide. Women account for 11 percent of computer scientists with Ph.D.s, and only 4 percent of full professors.

Roberts said there has been another disturbing trend over the last 10 years - a regression for women in the field. "There was a ramping- up of women's degrees in computer science until the mid '80s. But since that time there has been a decline in the fraction of women pursuing bachelor's degrees,” he said.

Computer science is losing women at the same time that the number of women getting degrees in other science and engineering fields has gone up, he said. Nationwide, women with bachelor's degrees in computer science presently number about 5,500 a year. This figure is less than half the number in the peak year, 1986.

“That reduction in the number of women entering the field and moving on to higher degrees makes it harder, given the pipeline attrition, to produce good computer scientists for industry and for academia,” Roberts said.

At Stanford over the last five years, 17 percent of the bachelor's degrees in computer science have gone to women. The 33- member computer science faculty includes three women, approximately 9 percent of the total.

Roberts said that he draws his knowledge of academic barriers, and how they are maintained, from personal experience while teaching at Wellesley and as a graduate student advocate for women at Harvard.

He also based his comments on more than a decade of research by scholars on women's progress in computing and other scientific fields.

He told the audience that his awareness of the barriers that women face was derived from three sources: the influence of the women in his family, specifically his grandmother and an aunt; the political culture of the 1960s when he grew up; and his involvement with the feminist community in Cambridge during the 1970s.

Thanks to these influences, after he received his doctorate in applied mathematics from Harvard, Roberts applied to teach only at women's colleges.

At Wellesley, where he taught from 1980 to 1984, Roberts' activism took the form of helping to lay the foundations for a computer science department.

“I know I made a statistical difference at Wellesley,” he said. "In 1984 Wellesley graduated more women in computer science than Stanford has graduated in any single year."

At Stanford, Roberts was named a Bing Fellow in 1993 in recognition of his excellence in teaching. He now serves as associate chair for education in the Department of Computer Science.

Two sets of hurdles

Women in computer science must overcome two sets of obstacles to success, Roberts said. Some are impediments in most scientific and technical fields:

  • Overt sexism, unwanted attention and sexual harassment create hostile working conditions.
  • A lack of role models for women in technical fields is discouraging. "When faculty members are looking for the next person to win a Turing Award, which is computer science's Nobel Prize, they tend to look for people like the last ones who won such awards. This usually involves looking in the mirror,” Roberts said.
  • The number of women in technical fields is not large enough to establish a resilient, supportive peer community.
  • Failures in the mentoring process contribute to women's lack of interest. Men advance academically because the faculty push them to do research and excel.
  • Expectations are different for male and female students. "Many women find that they get pats on the head for merely doing well in courses,” he said. As a result, they are not adequately prepared to engage challenges at higher stages of the academic pipeline.

Women who enter computer science must overcome an additional set of hurdles:

  • Experience in computer use prior to college differs markedly with gender. Boys get introduced to computers, and encouraged to use them, at an earlier age than girls. Girls get discouraged from using computers. Later, this prior experience privileges the men who enter computer-related fields.

In other technical fields, prior exposure tends to be more balanced, Roberts said.

  • Programming requires a mental state - something close to obsession - that Roberts said has been discouraged in women.

For instance, a programmer cannot say, “I will finish writing this code tomorrow.” By the next day, the depth and understanding that had been gained the day before will have been lost.

  • “Although it is by no means universal, there is some truth in the stereotypical images attached to programming and programmers,” Roberts said.

The image of computer scientists is one of nerds and hackers - people without social lives. “Women are better inculcated with social skills,” he said. They shy away from the image of a computer scientist, and thus reject computer science as a career option.

  • "Because software is so flexible, it allows the cultural imprinting of gender to permeate its design in a very interesting and insidious way," Roberts said. For instance, since most video games are written by men and reflect their interests, most have a level of violence that is not very interesting to young women.

Suggestions for improvement

Roberts offered several recommendations to increase the number of female computer scientists in academia. One is to expand existing support programs such as Women in Computer Science (WICS).

Another is to design attractive and exciting introductory courses in computer science, as well as structuring the entire undergraduate curriculum in a way that is challenging, yet accessible.

"The place that is most important to concentrate is the undergraduate years,” said Roberts. Once women have survived the weeding-out stages at the undergraduate level, then they will become solidly grounded in the basics of the field. Those whose interest is sustained in those critical years can count on "inertia" to carry them through post-graduate years.

He said women also should be provided with the resources to network not just locally but worldwide in order to build a strong support community.

And as computer science departments work to increase research opportunities, he said they also should encourage women to participate in that research.

Finally, he recommended that institutions explore alternative teaching approaches for computer scientists.

Some writers have argued that some aspects of methodology of the programming, such as top-down design and hierarchical decomposition, reflect male-dominated culture and therefore might be an unsuitable means of learning for women, Roberts said.

But he said there is a danger to using different means of teaching programming.

"Arguing that women do better using a different paradigm is easily interpreted as a tacit admission that they cannot succeed under the conventional approach," Roberts said.

Women who are taught under a different programming methodology will be less likely to work on projects using traditional methods. This could limit their power and participation.

"You can't do programming in a vacuum. It's a collective enterprise," he said. Everyone has to have the same foundation and the same knowledge.

On the other hand, one major advantage of computer science is that it is a growing and highly competitive field . Women who make it past the obstacles and obtain degrees in the field have a good chance of getting jobs.

Research reports and general information on women in computer science are available on the World Wide Web: and

The Jing Lyman Lecture Series is sponsored by the Institute for Research on Women and Gender.



This is an archived release.

This release is not available in any other form. Images mentioned in this release are not available online.
Stanford News Service has an extensive library of images, some of which may be available to you online. Direct your request by EMail to

© Stanford University. All Rights Reserved. Stanford, CA 94305. (650) 723-2300.